Tests · depression
Depression Test (PHQ-9)
The PHQ-9 helps you see how much low mood and low energy are affecting your everyday life right now. It's nine simple questions about the past two weeks, and together they give you a clear picture of where you stand. Your result can help you decide what to do next: try some self-support exercises or reach out for help.
- 9 questions
- Kroenke et al., 2001
- About this test
About the PHQ-9
How often nine common symptoms of depression, including thoughts of death or self-harm, bothered you over the last two weeks, added into one score.
- Developed by
- Robert Spitzer, Janet Williams and Kurt Kroenke and colleagues; validation paper published in 2001
- Built and tested in
- US primary care and obstetrics-gynecology clinics
Where it comes from
Robert Spitzer, Janet Williams and Kurt Kroenke and their colleagues built the Patient Health Questionnaire (PHQ) as a self-administered version of PRIME-MD, a diagnostic tool for common mental disorders in primary care. The PHQ-9 is its depression module. An educational grant from Pfizer supported the work. Its nine items are the nine DSM-IV symptoms of major depression, each scored from 0 (not at all) to 3 (nearly every day).
Kroenke, Spitzer and Williams published the validation paper in the Journal of General Internal Medicine in 2001. They wanted to test whether the same nine items could also grade how severe depression is, since severity helps guide treatment decisions.
What the research found
The paper drew on 6,000 patients: 3,000 from eight primary care clinics (mean age 46, 66% women, 79% white) and 3,000 from seven obstetrics-gynecology clinics. Reliability was high (Cronbach's alpha 0.89 in the primary care group and 0.86 in the other). As scores rose, patients reported worse functioning on all six domains of a general health survey, more sick days and more clinic visits.
In 580 patients who also had a telephone interview with a mental health professional, a score of 10 or higher had 88% sensitivity and 88% specificity for major depression. Only 41 of the 580 had major depression, so the positive predictive value ranged from 31% at a cut point of 9 to 51% at 15. The authors called scores from 10 to 14 a gray zone.
A 2019 analysis of individual participant data from 58 studies (17,357 participants) found the best combined accuracy at a cut-off of 10 or higher. In the 29 studies that used semistructured interviews, sensitivity was 0.88 and specificity 0.85. A 2012 meta-analysis of 18 studies (7180 people) found no substantial difference in accuracy for cut-offs from 8 to 11.
How scoring works
| Score | Label | What it means |
|---|---|---|
| 0 to 4 | Minimal signs | The authors' table calls this none to minimal depression and suggests no treatment action. |
| 5 to 9 | Mild low mood | The authors call this mild depression. Their clinical table suggests watchful waiting and repeating the questionnaire at follow-up. |
| 10 to 14 | Moderate low mood | The authors call this moderate depression. Their table suggests a clinician consider a treatment plan, such as counseling, follow-up or medication. A score of 10 is also the usual screening cut-off. |
| 15 to 19 | Moderately severe low mood | The authors call this moderately severe depression and suggest active treatment with medication, psychotherapy or both. |
| 20 to 27 | Severe | The authors call this severe depression. Their table, written for clinicians, suggests prompt treatment and, if impairment is severe or treatment isn't working, quick referral to a mental health specialist. |
The cut points of 5, 10, 15 and 20 are the authors'. The 2001 paper says they were chosen partly because they're simple for clinicians to remember, and that other cut points didn't noticeably change how severity related to function and health care use. It reported sensitivity and specificity against interviews for cut-offs from 9 to 15. The authors' own labels for 15 to 19 and 20 to 27 are "moderately severe" and "severe". The treatment suggestions come from a table written for clinicians, not for people taking the test on their own.
What it can't tell you
The PHQ-9 adds up nine symptoms and can't say what's behind them. Its authors expect a clinician to rule out physical causes, normal bereavement and a history of mania before making a diagnosis of depression, and a score can't do that. The 2001 study checked the scale against telephone interviews in US clinics, and the authors named its cross-sectional design as the main limit. The 2019 pooled analysis found higher sensitivity against semistructured interviews than against fully structured ones, and specificity that may be lower in younger people.
Item 9 asks about thoughts of being better off dead or of hurting yourself. The developers' manual says a final decision about the actual risk of self-harm requires a clinical interview. In a study of 84,418 outpatients, the one-year risk of a suicide attempt rose from 0.4% among people who answered "not at all" to 4% among those who answered "nearly every day." That's a pattern across a large group, and it doesn't predict what any one person will do.
How to use your result
A score describes the last two weeks, not who you are. The authors' levels start at 5 (mild), 10 (moderate), 15 (moderately severe) and 20 (severe). A score of 10 or higher is a good reason to talk with a doctor or therapist, who can ask about causes and how long this has gone on. A lower score doesn't rule out a hard stretch, so if you feel worse than the number says, tell someone who can help.
If you marked anything above "not at all" on the last question, please reach out now. Call or text 988 (the 988 Suicide & Crisis Lifeline), which is free and open 24/7, or call 911 if you're in immediate danger. For low mood, see behavioral activation and what helps when you can't get out of bed.
More on this topic: Behavioral activation, When you can't get out of bed, Anxiety test (GAD-7), Stress test (PSS-10).
Your answers
Your answers are scored in your browser. If you sign in and choose to save a result, we store the score and its label, not your individual answers. Without an account, nothing is saved.
Sources 9
- Kroenke, Spitzer and Williams, Journal of General Internal Medicine, 2001 (PubMed Central): the full article: sample and settings, scoring, reliability, sensitivity and specificity at each cut point, positive predictive value, the 5/10/15/20 cut points and why they were chosen, the item 9 rule, stated limits.
- Same paper on PubMed (abstract): abstract only: 6,000 patients, 580 interviews, sensitivity 88% and specificity 88% at 10 or higher.
- Levis, Benedetti and Thombs, BMJ, 2019 (PubMed): abstract only: individual participant data meta-analysis of 58 studies, cut-off 10, difference by interview type and age.
- Manea, Gilbody and McMillan, CMAJ, 2012 (PubMed): abstract only: meta-analysis of 18 studies, acceptable accuracy for cut-offs from 8 to 11.
- Simon et al., Psychiatric Services, 2013 (PubMed): abstract only: item 9 answers and later suicide attempts in 84,418 outpatients.
- PHQ Screeners instruction manual: the developers' bands and treatment table, the statement that item 9 needs a clinical interview, and the public domain statement.
- PHQ-9 form on phqscreeners.com: official item wording and the line that no permission is required to reproduce, translate, display or distribute.
- PHQ Screeners terms of use: the note that a depression diagnosis also requires ruling out bereavement, mania and physical or drug causes, and that the screeners are free for download and use as stated on the site.
- 988 Suicide and Crisis Lifeline: call or text 988, free, confidential, 24/7.